Learn Before
Primary Decisions for Non-End-to-End Pipelines
Question: What are the two core choices that must be resolved when designing a non-end-to-end machine learning pipeline?
Sample answer: The two core choices are deciding what steps the pipeline will contain and deciding how those steps connect or plug together.
Key points:
- Deciding the steps of the pipeline
- Deciding how the steps connect or plug together
Rubric: Answers must identify deciding what steps are in the pipeline and how they connect or plug together.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
Task Simplicity for Pipeline Component Selection
When designing a non-end-to-end pipeline, which two decisions must a practitioner make?
Choosing not to use an end-to-end system eliminates the need to make architectural decisions about your ML system.
If you choose not to use an end-to-end system, you must decide what steps are in your _____ and how they should plug together.
Match each pipeline design term to its correct description.
Arrange the design activities a practitioner follows when building a non-end-to-end ML pipeline.
In ML Yearning, what does deciding 'how steps plug together' primarily refer to in a non-end-to-end pipeline?
Designing a non-end-to-end pipeline requires deciding both what steps to include and how those steps connect to each other.
According to ML Yearning, after deciding what steps a pipeline contains, you must determine how they should _____ together.
Match each pipeline design scenario to the specific decision it illustrates.
Order the reasoning steps a practitioner uses when deciding whether and how to design a non-end-to-end pipeline.
Core Decisions in Non-End-to-End Pipeline Design
Structuring an Alternative Pipeline System
Primary Decisions for Non-End-to-End Pipelines